americant nov itates -...

16
AMERICANt MUSEUM Nov itates PUBLISHED BY THE AMERICAN MUSEUM OF NATURAL HISTORY CENTRAL PARK WEST AT 79TH STREET, NEW YORK, N.Y. 10024 Number 3202, 15 pp., 5 figures, 1 table August 29, 1997 A New Species of Cryptotis (Insectivora: Soricidae) from Northern Peru ELENA VIVAR,1 VICTOR PACHECO,2 AND MICHAEL VALQUI3 ABSTRACT The South American small-eared shrews of the genus Cryptotis constitute a poorly known group in need of much systematic revision. These shrews oc- cur in the Andes of Venezuela, Colombia, Ecuador, and northern Peru. They are informally recognized as the thomasi complex group (sensu Choate, 1970), which includes five species: C. avia, C. meridensis, C. montivagus, C. squamipes, and C. thomasi (sensu Hutterer, 1993). This paper reports a new species of the genus Cryptotis from Peru and provides an in- troduction to the systematic status of related taxa. Also, we recognize C. equatoris (including osgoodi) and C. medellinius as valid species distinct from C. thomasi. RESUMEN Las musaranias sudamericanas del genero Cryp- totis conforman un grupo taxon6mico pobremente conocido en necesidad de una exhaustiva revisi6n sistemaitica. Estas musaranias ocurren en los Andes de Venezuela, Colombia, Ecuador y norte del Peru, y son informalmente reconocidas como miembros del grupo complejo thomasi (sensu Choate, 1970). Este grupo incluye las siguientes especies: C. avia, C. meridensis, C. montivagus, C. squamipes y C. thomasi (sensu Hutterer, 1993). Este articulo reporta una nueva especie del genero Cryptotis del Peru' y provee una introduccion a la revisi6n sistemaitica de taxa relacionados. Basado en nuestros resultados, reconocemos tambien a C. equatoris (que incluye osgoodi) y C. medellinius como especies vailidas y distintas de C. thomasi. I Research Associate, Museo de Historia Natural, Universidad Nacional Mayor de San Marcos, Apartado 14-0434, Lima-14, Peru. 2Graduate Student, Department of Mammalogy, American Museum of Natural History; Biology Department, The City College of CUNY, New York; Curator of Mammals, Museo de Historia Natural, Universidad Nacional Mayor de San Marcos, Lima, Peru. 3Graduate Student, Department of Wildlife Ecology and Conservation, University of Florida, Gainesville, FL 32611. Copyright © American Museum of Natural History 1997 ISSN 0003-0082 / Price $2.10

Upload: lyque

Post on 09-Nov-2018

213 views

Category:

Documents


0 download

TRANSCRIPT

AMERICANt MUSEUM

Nov itatesPUBLISHED BY THE AMERICAN MUSEUM OF NATURAL HISTORYCENTRAL PARK WEST AT 79TH STREET, NEW YORK, N.Y. 10024

Number 3202, 15 pp., 5 figures, 1 table August 29, 1997

A New Species of Cryptotis(Insectivora: Soricidae) from Northern Peru

ELENA VIVAR,1 VICTOR PACHECO,2 AND MICHAEL VALQUI3

ABSTRACT

The South American small-eared shrews of thegenus Cryptotis constitute a poorly known group inneed of much systematic revision. These shrews oc-cur in the Andes of Venezuela, Colombia, Ecuador,and northern Peru. They are informally recognizedas the thomasi complex group (sensu Choate, 1970),which includes five species: C. avia, C. meridensis,

C. montivagus, C. squamipes, and C. thomasi (sensuHutterer, 1993). This paper reports a new species ofthe genus Cryptotis from Peru and provides an in-troduction to the systematic status of related taxa.Also, we recognize C. equatoris (including osgoodi)and C. medellinius as valid species distinct from C.thomasi.

RESUMEN

Las musaranias sudamericanas del genero Cryp-totis conforman un grupo taxon6mico pobrementeconocido en necesidad de una exhaustiva revisi6nsistemaitica. Estas musaranias ocurren en los Andesde Venezuela, Colombia, Ecuador y norte delPeru, y son informalmente reconocidas comomiembros del grupo complejo thomasi (sensuChoate, 1970). Este grupo incluye las siguientes

especies: C. avia, C. meridensis, C. montivagus,C. squamipes y C. thomasi (sensu Hutterer, 1993).Este articulo reporta una nueva especie del generoCryptotis del Peru' y provee una introduccion a larevisi6n sistemaitica de taxa relacionados. Basadoen nuestros resultados, reconocemos tambien a C.equatoris (que incluye osgoodi) y C. medelliniuscomo especies vailidas y distintas de C. thomasi.

I Research Associate, Museo de Historia Natural, Universidad Nacional Mayor de San Marcos, Apartado 14-0434,Lima-14, Peru.

2Graduate Student, Department of Mammalogy, American Museum of Natural History; Biology Department, TheCity College of CUNY, New York; Curator of Mammals, Museo de Historia Natural, Universidad Nacional Mayorde San Marcos, Lima, Peru.

3Graduate Student, Department of Wildlife Ecology and Conservation, University of Florida, Gainesville, FL32611.

Copyright © American Museum of Natural History 1997 ISSN 0003-0082 / Price $2.10

AMERICAN MUSEUM NOVITATES

INTRODUCTION

The small-eared shrews of the genus Cryp-totis are among the poorest known taxa inthe family Soricidae. They are widely dis-tributed in the Americas, ranging from partsof North America, Middle America to north-western South America. Cryptotis is the onlyinsectivore with a distribution reaching SouthAmerica. These shrews occur in cloud forestsand paramo habitats in the Andes of Vene-zuela, Colombia, Ecuador, and northern Peru,frequently above 1200 m elevation (Nowak,1991; Hutterer, 1993). The taxonomy andphylogenetic relationships within the genusare obscure, and although some progress hasbeen made, their systematics is still far fromresolved. Choate (1970), in a partial revisionof the genus, recognized four lineage groups:the Cryptotis mexicanus group including thespecies C. mexicanus, C. goldmani, and C.goodwini; the Cryptotis parvus group includ-ing C. parvus and C. nigrescens; the Cryp-totis thomasi group including all the SouthAmerican taxa known at that time; and a rel-ict species group that included Middle Amer-ican taxa of uncertain allocation. A more for-mal characterization of Choate's thomasispecies group has been elusive, impededlargely by the lack of alpha-systematic stud-ies.A more comprehensive systematic review

has been presented for Middle Americanforms. Woodman and Timm (1992) de-scribed a new species, Cryptotis hondurensis,from Honduras. They later recognized theCryptotis nigrescens complex group to in-clude all taxa previously known as nigres-cens, resurrecting the species mayensis, mer-riami, and merus. They also stated that Cryp-totis hondurensis tnight be embedded in theC. nigrescens group (Woodman and Timm,1993).Our knowledge of the systematics of

South American shrews is incipient. Hutter-er's (1993) tentative arrangement providedfor Cryptotis recognized the following SouthAmerican species: C. aviua G. M. Allen,1923; C. meridensis Thomas, 1898; C. non-tivagus (Anthony, 1921); C. squamipes (J. A.Allen, 1912); and C. thomasi (Merriam1897). In addition, Woodman and Timm(1993) described C. colombianus from the

Central Cordillera of Colombia, relating it tothe C. nigrescens species complex, whichwould indicate that South American shrewspecies do not make a monophyletic group(our interpretation). And recently, Woodman(1996) considered avia a junior synonym ofthomasi.

Cabrera (1958) and Hutterer (1993) bothconsidered Cryptotis medellinius Thomas,1921, C. equatoris (Thomas, 1912), and C.osgoodi (Stone, 1914) synonyms of Cryptotisthomasi. It appears that only Anthony (1921)and Tate (1932) suggested that equatoris andosgoodi might be conspecific, and distinctfrom thomasi. Moreover, Tate (1932) consid-ered medellinius a distinct species, differentfrom thomasi; but no morphological com-parisons to support his arrangement wereprovided.

Until recently, the genus Cryptotis was un-known from Peru (Honacki et al., 1982; Cor-bet and Hill, 1991). However, its occurrencein northern Peru was reported in a generalgeographic overview of the country (Brack-Egg, 1986), a record that is undoubtedlybased on the single specimen collected byLinda Barkley in Piura, Peru. To the best ofour knowledge, this specimen was neverstudied and reported, except by Hutterer(1993) who included it as Cryptotis thomasifor northern Peru. In 1992, the Centro de Da-tos para la Conservacion de la UniversidadAgraria La Molina conducted an inventoryof the paramo and montane forest of San Ig-nacio and Ja6n, Cajamarca, northern Peru.This inventory produced an interesting set ofsmall mammals including a small-earedshrew found dead along a peasant trail. Thisshrew resembles Barkley's specimen; bothexhibit peculiar characters that differentiatethem from other known taxa. We describethese two specimens as evidence of a newspecies and discuss the status and content ofrelated taxa. We also hope this paper en-courages a thorough revision of the SouthAmerican small-eared shrews.

ACKNOWLEDGMENTS

We acknowledge the following curatorsfor allowing use of specimens for compari-sons and illustrations: M. Haffner (LSUMZ),

2 NO. 3202

VIVAR ET AL.: NEW SPECIES OF CRYPTOTIS

M. Moreno (MECN), and B. D. Patterson(FMNH). We thank R. Hutterer who kindlyguided Vivar during the first steps of thiswork. Robin Foster's hospitality to Vivar inChicago is also greatly appreciated. Thework of Pacheco in New York was partiallyfinanced by the Office of Grants and Fellow-ships of the AMNH. We acknowledge theCentro de Datos para la Conservacion de laUniversidad Nacional Agraria La Molina(CDC) for supporting the expedition that re-sulted in the collection of the holotype. B. D.Patterson and R. Hutterer kindly providedcomments on the manuscript. We also thankMs. Paula Jenkins of the British Museum ofNatural History for information on the ho-lotype of C. thomasi. Finally, we express ourgratitude to Maritza Ocrospoma for skillfullydrawing the cranial figures.

MATERIALS AND METHODS

Skin measurements were taken from la-bels. The head and body length (HBL) wasestimated by subtracting the tail length (TL)from the total length. All cranial and man-dibular variables were measured to the near-est 0.1 mm using dial calipers. Most of thevariables used here follow and are illustratedin Woodman and Timm (1993) and Wood-man (1995). The nomenclature for dental to-pography follows Choate (1970). The fol-lowing craniomandibular measurements wereused: condylobasal length, not including theupper incisors (CBL); cranial breadth (CB);breadth of zygomatic plate (ZP); interorbitalbreadth (10); breadth of palate across firstunicuspid (U1B); breadth of palate acrossthird unicuspid (U3B); breadth of palateacross second molars (M2B); palatal length(PL); upper tooth row length, UI to M3, par-allel to the long axis of the skull (TRL); un-icuspid tooth row length (UNL); posteriorwidth of Ml across hypocone and metastyle(WM 1); mandibular length from inferior sig-moid notch to posterior edge of mental fo-ramen (ML); height of coronoid process(HCP); height of coronoid valley (HCV);lower tooth row length, p3 to m3 (LTL);height of articular condyle (HAC); and widthof articular condyle (WAC). The last twovariables follow Woodman (1995).

Specimens are housed at the Field Muse-

um of Natural History, Chicago (FMNH);the American Museum of Natural History,New York (AMNH); Louisiana State Uni-versity Museum of Zoology, Baton Rouge(LSUMZ); Museo Ecuatoriano de CienciasNaturales, Quito (MECN); and Museo deHistoria Natural de la Universidad NacionalMayor de San Marcos, Lima (MUSM).

Cryptotis thomasi (Merriam, 1897)Figure 1

Blarina thomasi Merriam, 1897: 227.

HOLOTYPE: Female. BM(NH) 97.5.21.2."Plains of Bogota, Colombia (on G. D.Child's Estate, near city of Bogota, alt. about9000 ft)." The original description states thatit was collected on 14 November 1895, butapparently it was collected on 14 September1895 (P. Jenkins, personal comun.).

DISTRIBUTrIoN: Paramos of Bogota, Colom-bia.

DESCRIPTION: A large-size Cryptotis, HBLaveraging 86 (table 1); tail very short, aver-aging 30% of HBL; dorsal and lateral pelagelight brown; ventral pelage slightly paler.

Rostrum long and narrow; nasal openingnarrow; interorbital area narrow; braincasenot notably inflated; two small dorsal foram-ina positioned anteriorly to the dorsal artic-ular facet; anterior process of the petromas-toid high and moderately wide; large fora-men on posterior edge of tympanic processof petromastoid; paroccipital process low;zygomatic plate narrow; anterior border ofzygomatic plate above or slightly in front ofmetastyle of MI, and posterior border aboveor slightly behind M2 and M3 border; palatenarrow; U4 medium size, reduced to abouthalf the size of U3; U4 in line with otherunicuspids, preventing contact between U3and P4; U4 partially obscured by P4 but stillvisible from lateral view; posterior borders ofP4, Ml, and M2 slightly recessed; M3 com-plex: protocone, paracone, parastyle, and me-sostyle conspicuous; metacone reduced andhypocone absent or part of posterior cingu-lum; upper dentition semibulbous.

Mandible long and narrow; coronoid pro-cess low and narrow, joining mandible atoblique angle; coronoid process flared to la-bial side of mandible; articular process dis-tinctive, high and broad with a lingual notch

1997 3

AMERICAN MUSEUM NOVITATES

Fig. 1. Dorsal, ventral, and lateral views of skull, and lateral(FMNH 71037).

between the two articular facets; lower sig-moid notch slightly deep, below ventral ar-ticular facet; angular process long and slen-der; posterior border of lower incisor slightlyanterior to posterior cingulum of pm4; hy-poconid and entoconid in talonid of m3. En-toconid is sometimes obscured because ofwear.

Cryptotis medellinius Thomas, 1921Figure 2

Cryptotis medellinius Thomas, 1921: 354.

HOLOTYPE: "Adult male. B. M. 21.7.1.9.Original number 10. Collected December1919 by Frere Niceforo Maria," from SanPedro, 30 km north of Medellin, Colombia.

DISTRIBUTION: Medellin region, Colombia.DESCRIPTION: A large-size Cryptotis, HBL

averaging 84 (table 1); tail medium-size, av-eraging 47% of HBL; dorsal and lateral pel-age light brown; ventral pelage slightly paler.

Rostrum long and relatively broad; nasalopening moderately broad; interorbital areabroad; braincase inflated; two small dorsalforamina positioned anterior to dorsal artic-

view of lower jaw, Cryptotis thomasi

ular facet; anterior process of petromastoidmoderately high and wide; foramen on pos-terior edge of tympanic process of petromas-toid; paroccipital process low; zygomaticplate narrow; anterior border of zygomaticplate at metacrista of MI and posterior bor-der above metacrista or metastyle of M2; pal-ate broad; unicuspid U4 usually small andslightly lingually placed preventing contactbetween U3 and P4; U4 barely visible fromlateral view; posterior borders of P4, Ml, andM2 slightly recessed; M3 complex: proto-cone, paracone, parastyle, and mesostyleconspicuous; metacone reduced and hypo-cone absent or part of posterior cingulum;dentition bulbous.

Mandible long and narrow; coronoid pro-cess moderately high and narrow joiningmandible at oblique angle; articular processhigh and broad; lingual notch present be-tween two articular facets of articular pro-cess; lower sigmoid notch slightly deep; an-gular process long and slender; posterior bor-der of lower incisor notably anterior to pos-terior cingulum of pm4; hypoconid in talonidof m3 reduced.

NO. 32024

VIVAR ET AL.: NEW SPECIES OF CRYPTOTIS

Fig. 2. Dorsal, ventral, and lateral views of skull, and lateral view of lower jaw, Cryptotis medellinius(AMNH 149151).

COMPARISONS: Compared to Cryptotis tho-masi, C. medellinius is about the same sizebut has a longer tail (47% of HBL vs. 30%);rostrum slightly broader; braincase more in-flated; palate broader; upper dentition morebulbous; coronoid process higher and ento-conid in talonid of m3 is absent.

REMARKS: The proper gender of mammaliangeneric names ending in -otis, the case ofCryptotis, has been subject of recent debate.Woodman (1993) has recently argued that allgeneric names ending in -otis are feminine,therefore recommending that species names ofall mammalian genera ending in -otis must beproperly corrected to agree in gender with thegeneric names (Art. 31b-Ride et al., 1985).Thus among other species, two South Ameri-can shrews were corrected: medellinia, andmontivaga (Woodman, 1993). The specificname avia is a noun in apposition and there-fore retains its original spelling (Woodman,1993). However, Pritchard (1994) contestedthat the ending -otis is a Latin derivation of theGreek term otos and that Latin nouns endingin -is are of the 3rd declension, and may be

masculine, feminine, or neuter, depending onpriority of usage. As Woodman (1993) andPritchard (1994) pointed out, the first specificname combination requiring gender agreementwith Cryptotis appears to be C. merus byGoldman (1912; assuming masculine gender).Steppan (1995) retained the historical usage ofmasculine specific names for the muroid Phyl-lotis, but suggested that this issue needs furtherinvestigation on the etymology of the Greekterms. Following Pritchard (1994) we maintainthe usage of medellinius, and montivagus, andspell out colombianus (for colombiana). AsPritchard (1994) pointed out, no other mam-malian generic names listed by Woodman(1993) that end in -otis have been combinedwith specific feminine epithets.

Cryptotis equatoris (Thomas, 1912)Figure 3

Blarina equatoris Thomas, 1912: 409.

HOLOTYPE: "Old male. B.M. 99.9.9.3.Original number 136. Collected 17th Decem-ber 1898 by Perry 0. Simons." Type from

1997

AMERICAN MUSEUM NOVITATES

Fig. 3. Dorsal, ventral, and lateral views of skull, and lateral view of lower jaw, Cryptotis equatoris(AMNH 66845).

Sinche, Guabanda [Guaranda], 4000 m, Ec-uador.

DISTRIBUTION: Central Andes of Ecuador.DESCRIPTION: A medium-size Cryptotis,

HBL averaging 79 (table 1); tail medium, av-

eraging 45% of HBL; dorsal and lateral pel-age dark brown; ventral pelage slightly paler.

Rostrum narrow and moderately long; na-

sal openings narrow; interorbital area nar-

row; braincase moderately inflated; two me-

dium size to small dorsal foramina anteriorto dorsal articular facet; anterior process ofpetromastoid low and narrow; posterior edgeof tympanic process lacks a foramen; par-

occipital process low; zygomatic plate nar-

row; anterior border of zygomatic plate atmetacrista slightly anterior to metastyle ofMl and posterior border behind the M2 andM3 border; palate narrow and moderatelyshort; U4 usually small, reduced to less thanhalf the size of U3; U4 in line with otherunicuspids which, although small, still pre-vent contact between U3 and P4; U4 ob-scured by P4 but still visible from lateralview; posterior borders of P4, MI, and M2

slightly recessed; M3 complex and well de-veloped: protocone, paracone, parastyle, andmesostyle conspicuous; metacone reducedbut fairly visible and hypocone absent or partof posterior cingulum; dentition semibul-bous.

Mandible long and narrow; coronoid pro-cess low and narrow joining mandible atoblique angle; coronoid process flared to la-bial side; articular condyle distinctive, low,and narrow; lingual notch between two artic-ular facets faint in some specimens; lowersigmoid notch slightly deep; posterior borderof lower incisor anterior to posterior cingu-lum of pm4; only hypoconid in talonid ofm3.

COMPARISONS: Compared to Cryptotis tho-masi, C. e. equatoris has a much darker dor-sal and ventral pelage; it is smaller (HBL=79 vs. 86) and has a longer tail (45% ofHBL vs. 30%); skull length dimensionssmaller on average; anterior process of petro-mastoid lower and narrower; posterior edgeof tympanic process of petromastoid lacks aforamen; U4 usually smaller; coronoid valley

NO. 32026

VIVAR ET AL.: NEW SPECIES OF CRYPTOTIS

lower; articular process lower and narrower;lingual notch between articular facets faint inseveral specimens; entoconid in talonid ofm3 absent.

REMARKS: Thomas (1912) described Cryp-totis equatoris from Sinche, Guabanda[Guaranda], Ecuador, mentioning that thehabitat included "other specimens from Mt.Pichincha, Quito." Later, Stone (1914) de-scribed C. osgoodi based on specimens col-lected at Hacienda Garzon on Mt. Pichincha.Anthony (1921) and Tate (1932) studied bothtaxa, and although no character or meristiccomparison was provided, each suggestedthat both taxa might be conspecific, thereforeosgoodi ought to be synonymized underequatoris based on seniority. Cabrera (1958)synonymized osgoodi and equatoris underCryptotis thomasi. The same arrangementwas followed by Hutterer (1993).

In the course of this work, we have foundsome differences between Sinche and Pichin-cha populations in scored characters (table1). The topotypic osgoodi compared to to-potypic equatoris is, on average, smaller inHBL, CBL, ML; narrower in CB, 10, U1B,U3B, M2B; and lower in HCP, indicating asmaller skull and narrower rostrum; whichmight suggest a species recognition. How-ever, various degrees of overlap are present.Pending studies of morphological charactersthat include intermediate populations and di-rect examination of holotypes, we concludethat osgoodi and equatoris are conspecific,concurring with Anthony (1921) and Tate(1932). However, based on the meristic dif-ferences, we believe it is reasonable to con-sider Cryptotis equatoris osgoodi a subspe-cies.

Cryptotis montivagus (Anthony, 1921)Blarina montivaga Anthony, 1921: 5.

HOLOTYPE: Adult female, American Mu-seum of Natural History (AMNH 47200)collected on 15 January 1921 by H. E. An-thony at Bestion, Prov. del Azuay, Ecuador,altitude 10,000 ft. Original number 2534.

DISTRIBUTION: Andes of Ecuador.DESCRIPTION: A large-size Cryptotis, HBL

averaging 85 (table 1); tail short, averaging37% of HBL; dorsal and lateral pelage lightgrayish brown; ventral pelage slightly paler.

Rostrum narrow and moderately long; na-sal openings narrow; interorbital area nar-row; braincase slightly inflated; dorsal fo-ramina fused or tightly placed anterior to thedorsal articular facet; anterior process of thepetromastoid low and narrow; posterior edgeof the tympanic process of the petromastoidlacks a foramen; paroccipital process low;zygomatic plate narrow; anterior border ofzygomatic plate vertically oriented and abovemetacrista of Ml; posterior border above theM2/M3 contact; palate narrow; U4 of medi-um size, reduced to about half the size of U3;U4 in line with other unicuspids, preventingcontact between U3 and P4; U4 visible fromlateral view; posterior borders of P4, Ml, andM2 recessed; M3 complex: protocone, para-cone, parastyle, and mesostyle conspicuous;metacone reduced and hypocone absent orpart of posterior cingulum; dentition semi-bulbous.

Mandible long and narrow; coronoid pro-cess low and narrow, joining mandible atoblique angle; articular condyle distinctive,high and broad; lower sigmoid notch slightlydeep; posterior border of lower incisor an-terior to posterior cingulum of pm4; only hy-poconid in talonid of m3.

Cryptotis peruviensis, new speciesFigure 4

Cr[i]ptotis sp.: Brack-Egg, 1986: 301.Cryptotis thomasi: Hutterer, 1993: 109 [part].Cryptotis sp.: Pacheco et al. (1995: 7, 33).

HOLOTYPE: Subadult female, Museo deHistoria Natural de la Universidad NacionalMayor de San Marcos (MUSM 8373), col-lected by Antonio Tovar, and Michael andThomas Valqui on 19 February 1992, fieldnumber LATN 46. The holotype is a flat skinmounted on cardboard in good condition anda complete skull, mandibles, and partial skel-eton.

TYPE LOCALITY: Peru, Department Caja-marca, Las Ashitas, 3150 m, about 42 km Wof Jaen (05°42'S, 79°08'W).

PARATYPE: Adult male, LSUMZ 26887.The paratype is a skin, skull, and skeleton ingood condition; collected by Linda J. Bark-ley in "Machete" on Zapalache-Carmen trailat 2050 m, Department Piura on 19 June1980.

71997

AMERICAN MUSEUM NOVITATES

Fig. 4. Dorsal, ventral, and lateralviensis, new species (MUSM 8373).

DISTRIBUTION: Known only by two speci-mens; one from the eastern versant of theAndes of northern Peru, Department Caja-marca, the other from the western side of theAndes, Department Piura (fig. 5). These re-cords represent the southernmost distribu-tional limits for the genus. The species rangemost likely includes the southern edge of theAndes north of the Huancabamba Depres-sion.

DIAGNOSIS: A small-size species of Cryp-totis characterized by its dark grayish browncoloration; tail medium size (48% of headand body length); nasal openings broad; ros-trum long and narrow; braincase inflated; an-terior process of petromastoid low and mod-erately wide to narrow; upper incisors withbroad basal cusp and short slender hook;dentition bulbous; unicuspid cups with shal-low borders; unicuspid U4 small, largely pre-venting contact between U3 and PM4, in-conspicuous from lateral view; articular con-dyle distinctive, high, and broad.MEASUREMENTS OF HOLOTYPE: Measure-

views of skull, and lateral view of lower jaw, Cryptotis peru-

ments are in millimeters and weight ingrams; external measurements are those re-corded on the specimen label: head and bodylength 63; tail length 30.5; hind foot includ-ing claws 14; ear length (not measured);weight 9 g. See table 1 for additional mea-surements.

DESCRIPTION: Cryptotis peruviensis issmall compared to other South Americanspecies (HBL = 68 ± 5); tail of mediumlength averaging 48% of HBL; fore and hindfoot medium size, not specialized, coveredwith short hairs and small scales; dorsal andlateral pelage coloration dark grayish brown(color 20 of Smithe, 1975); ventral pelageslightly paler, without a clear demarcationbetween dorsum and venter; dorsal fur of ho-lotype is densely woolly, slightly less so inparatype; dorsal hairs approximately 5-6 mmlong.The rostrum is long (PL/CBL = 43, 45%)

and narrow (M2B/CB = 56, 58%); nasalopenings broad; interorbital area narrow;braincase inflated; two small dorsal foramina

8 NO. 3202

VIVAR ET AL.: NEW SPECIES OF CRYPTOTIS

positioned anteriorly to dorsal articular facet;anterior process of petromastoid low, mod-erately wide in holotype and narrow in para-type; foramen on posterior edge of tympanicprocess absent; paroccipital process low; zy-gomatic plate narrow; anterior border of zy-gomatic plate above metastyle of MI, pos-terior border above M3; palate narrow; upperincisor with broad basal cusp and slenderhook, both nearly at same level; unicuspidcusps with shallow borders; U4 small and tri-angular in occlusal view, in paratype U4even smaller, probably because of wear; U4in line with other unicuspids, does not pre-vent contact between U3 and P4, or nearlyso, in paratype; U4 not visible from lateralview; posterior borders of P4, MI, and M2slightly recessed; M3 complex, protocone,paracone, parastyle, and mesostyle conspic-uous, metacone reduced and hypocone ab-sent or part of posterior cingulum; dentitionbulbous with dark pigmentation in holotypeand slightly less in paratype.

Mandible long and narrow; coronoid pro-cess low and narrow joining mandible atoblique angle; coronoid process flared to la-bial side of mandible; articular condyle dis-tinctively high and broad, presenting a shal-low lingual notch between articular facets;lower sigmoid notch slightly deep, belowventral articular facet; angular process longand slender; lower incisor with three con-spicuous cusps; posterior border of lower in-cisor extends anterior to posterior cingulumof pm4; hypoconid in talonid of m3; a smallentoconid is visible in paratype.

COMPARISONS: Cryptotis peruviensis canbe readily distinguished from the followingSouth American taxa based on a combinationof size and other external characters: Cryp-totis meridensis, Cryptotis squamipes, andCryptotis colombianus. For descriptions ofthose taxa, readers are referred to Choate andFleharty (1974), Hutterer (1986, 1993), andWoodman and Timm (1992, 1993). Cryptotisperuviensis needs comparisons only with C.thomasi, other taxa previously included un-der that name, and with Cryptotis montiva-gus, which has a range close to that of C.peruviensis.

Compared to Cryptotis medellinius, C. pe-ruviensis is smaller in head and body averagelength (68 vs. 84), but has similar tail length

(48% of HBL vs. 47%); dorsal pelage isdarker and grayish; broader nasal opening;lacks a foramen on posterior edge of tym-panic process of petromastoid; narrower pal-ate; smaller U4, preventing contact betweenU3 and P4, and U4 not visible from labialview; dentition is more bulbous; lower cor-onoid process.Compared to Cryptotis thomasi, C. peru-

viensis is smaller (HBL = 68 vs. 86) and haslonger tail (48% of HBL vs. 30%); darkerdorsal pelage; less hair on fore and hind feet;broader nasal opening; relatively more in-flated braincase; lacks a foramen on tympan-ic process of petromastoid bone; dentitionbulbous and borders more shallow; smallerU4, preventing contact between U3 and P4or nearly so, and U4 not visible from labialview.Compared to Cryptotis e. equatoris, C. pe-

ruviensis is smaller in average head and bodylength (68 vs. 79 mm); has longer tail length(48% of HBL vs. 45%); narrower rostrum;broader nasal openings; more inflated brain-case; smaller U4 that prevents contact be-tween U3 and P4 or nearly so, and is notvisible from lateral view; dentition bulbous;higher coronoid valley; articular condyle pro-cess higher and broader.Compared to Cryptotis e. osgoodi, C. pe-

ruviensis, although of similar size in HBLand TL, has broader nasal openings; moreinflated braincase; broader rostrum; smallerU4 that prevents contact between U3 and P4or nearly so, and U4 not visible from lateralview; bulbous dentition; mandible slightlylonger and more robust; articular condyleprocess higher and broader.Compared to Cryptotis montivagus, C. pe-

ruviensis is smaller in average head and bodylength (68 vs. 85); has longer tail length(48% of HBL vs. 37%); broader nasal open-ings; broader interorbital area; inflated brain-case; shorter rostrum; shorter unicuspidlength; smaller U4 preventing contact be-tween U3 and P4 or nearly so, and U4 notvisible from lateral view; less recessed pos-terior border of molariforms; bulbous denti-tion; longer mandible.

HABITAT DESCRIPTION: Las Ashitas campwas located in an elfin forest habitat, char-acterized by shrubby trees with abundant ep-iphytes and moss, and the holotype was col-

91997

AMERICAN MUSEUM NOVITATES

TABLE 1Selected Measurements, South American Short-Eared Shrews of the Genus Cryptotis

C. peruviensis(holotype and

C. medellinius C. thomasi C. e. equatoris C. e. osgoodi C montivagus paratype)

HBL 83.8 ± 6.677.0-94.0

(4)TL 39.0 ± 3.7

34.0-43.0(4)

TLP 46.8 ± 5.343.0-55.8

(4)CBL 21.3 ± 0.5

21.0-22.0(3)

CB 10.8 ± 0.210.5-11

(3)ZP 2.1 ± 0.2

1.7-2.3(4)

10 5.2 ± 0.15.0-5.3

(4)U1B 2.8 ± 0.1

2.7-2.9(4)

U3B 3.3 ± 0.13.2-3.4

(4)M2B 6.4 ± 0.2

6.2-6.7(5)

PL 9.6 ± 0.29.4-9.9

(5)TRL 8.4 ± 0.1

8.3-8.6(5)

UNL 2.8 ± 0.12.6-3.0

(5)WM1 2.0 ± 0.1

1.8-2.1(5)

ML 7.3 ± 0.27.0-7.4

(5)HCP 4.9 + 0.2

4.7-5.2(5)

HCV 3.2 ± 0.22.9-3.5

(5)

86.3 ± 4.379.0-93.0

(6)25.5 ± 1.423.0-27.0

(6)29.6 ± 1.427.4-31.6

(6)21.6 ± 0.221.1-21.9

(7)10.5 ± 0.110.3-10.6

(5)2.0 ± 0.11.8-2.2

(9)5.0 ± 0.14.7-5.1

(9)2.7 ± 0.12.6-2.8

(9)3.1 ± 0.13.0-3.2

(8)6.1 ± 0.25.8-6.4

(8)9.2 ± 0.29.0-9.6

(9)8.2 ± 0.17.9-8.4

(9)2.6 ± 0.12.4-2.8

(8)1.9 ± 0.11.8-2.0

(9)7.1 ± 0.26.9-7.5

(9)4.6 ± 0.14.4-4.8

(8)3.2 ± 0.13.0-3.4

(9)

79.0 ± 1.777.0-82.0

(7)35.4 ± 2.133.0-40.0

(7)44.9 ± 3.041.5-51.3

(7)21.0 ± 0.320.5-21.6

(6)10.3 ± 0.29.9-10.6

(6)2.2 ± 0.12.1-2.4

(7)5.1 ± 0.14.9-5.3

(7)2.7 ± 0.12.5-2.8

(7)3.2 ± 0.13.0-3.3

(7)6.2 ± 0.15.9-6.3

(7)9.2 ± 0.28.9-9.5

(7)8.0 ± 0.27.8-8.2

(7)2.7 ± 0.12.5-2.8

(7)1.9 ± 0.11.8-2.0

(7)7.3 ± 0.17.0-7.4

(7)4.6 ± 0.24.3-4.8

(7)2.8 ± 0.12.7-2.9

(7)

73.3 ± 5.664.0-79.0

(4)30.8 ± 1.829.0-33.0

(4)42.3 ± 5.036.7-50.0

(4)20.7 ± 0.120.5-20.8

(5)9.9 + 0.09.9-9.9

(2)2.0 ± 0.11.8-2.2

(7)4.9 + 0.14.8-5.1

(7)2.4 ± 0.12.3-2.5

(6)2.9 + 0.12.7-3.1

(6)5.9 + 0.05.8-5.9

(6)9.1 ± 0.18.9-9.3

(7)7.9 ± 0.27.7-8.2

(7)2.8 ± 0.12.7-3.0

(6)1.9 ± 0.11.8-2.0

(7)7.0 ± 0.26.7-7.2

(7)4.2 ± 0.14.0-4.4

(7)2.8 ± 0.12.7-2.9

(7)

84.5 ± 2.181.0-86.0

(4)31.0 ± 1.928.0-33.0

(4)36.7 ± 2.432.6-38.4

(4)21.1 ± 0.320.8-21.4

(2)10.4 t 0.010.4-10.4

(1)2.0 ± 0.11.9-2.1

(4)4.8 ± 0.14.6-4.9

(4)2.7 ± 0.02.7-2.8

(3)3.2 ± 0.13.1-3.4

(4)6.1 ± 0.16.0-6.2

(4)9.2 ± 0.19.1-9.4

(4)8.1 _ 0.27.9-8.3

(4)3.0 ± 0.12.8-3.1

(4)1.9 _ 0.11.8-2.0

(4)7.0 ± 0.16.8-7.0

(4)4.6 t 0.04.5-4.6

(4)3.1 ± 0.12.9-3.2

(4)

63, 73

31, 35

48, 48

20.6, 21.7

10.5, 10.7

2.1, 2.1

5.1, 5.2

2.7, 2.8

3.1, 3.2

5.9, 6.2

8.8, 9.7

7.9, 8.3

2.6, 2.7

1.8, 1.9

7.4, 7.6

4.6, 4.7

3, 3

10 NO. 3202

VIVAR ET AL.: NEW SPECIES OF CRYPTOTIS

TABLE 1-(Continued)

C. peruviensis(holotype and

C. medellinius C. thomasi C. e. equatoris C. e. osgoodi C. montivagus paratype)

HAC 3.0 + 0.1 3.1 + 0.1 2.7 + 0.1 2.6 + 0.1 3.0 + 0.1 2.9, 2.92.8-3.1 3.0-3.3 2.4-2.8 2.4-2.8 2.9-3.1

(5) (9) (7) (7) (4)WAC 2.0 + 0.1 1.9 ± 0.1 1.7 + 0.1 1.7 ± 0.1 1.9 + 0.1 1.9, 1.9

1.8-2.1 1.7-2.0 1.6-1.9 1.5-1.8 1.8-2.0(5) (9) (7) (7) (4)

LTL 6.5 + 0.2 6.4 + 0.1 6.4 + 0.1 6.4 + 0.1 6.5 ± 0.1 6.3, 6.86.4-6.9 6.2-6.5 6.2-6.6 6.3-6.6 6.3-6.7

(5) (8) (7) (7) (4)

lected at the upper limit of the forest, belowthe paramo. This locality corresponds to theBosque Pluvial Montano Tropical accordingto the Holdridge Life Zone Classification(Tosi, 1960). Common shrub species in-cludes representatives of the genera Gynoxus(Asteraceae), Monnina (Polygalaceae), Ra-panea (Myrsinaceae), Miconia (Melastoma-taceae), and Escallonia (Grossulariaceae);Syphocampilus jelskii (Campanulaceae), treeferns of the genus Cyathea; and bamboos ofthe genus Chusquea (Poaceae), intermixedwith patches of grassland species of Cala-magostris and Festuca. The environment iscold and extremely humid.The Machete field locality is described as

an extensive and typical cloud forest withpresence of Podocarpus trees placed on eastslope of Cerro Chinguela, in the Rio Saman-iego valley (Parker et al., 1985).

DISCUSSION

Woodman and Timm (1993) found that theCryptotis nigrescens species complex ismore diverse and speciose than was previ-ously considered. Our recognition here ofCryptotis equatoris and C. medellinius andthe report of C. peruviensis give us reason-able confidence to suggest that a similar pic-ture of high diversity characterizes the SouthAmerican short-eared shrews. A comprehen-sive revision of the group is needed to de-velop this probable scenario.The inclusion of all the South American

Cryptotis as members of a single Cryptotisthomasi species group (sensu Choate, 1970),based on geographic grounds but without

convincing evidence of its monophyly, ap-pears a simplistic view at this time, and notlonger useful in a systematic context. A morerealistic C. thomasi group would includeonly Cryptotis thomasi, and tentatively C.medellinius.

Cryptotis colombianus was described byWoodman and Timm (1993) as belonging tothe Central American nigrescens complexgroup; we interpret this to mean that theSouth American taxa do not conform amonophyletic group. However, in the courseof this study, some of the diagnostic char-acters of C. colombianus that distinguish itfrom Central American species were foundin other South American species. For exam-ple: the foramen on the posterior edge of thetympanic process of the petromastoid waspresent in C. thomasi and C. medellinius; themoderately high and wide anterior process ofthe petromastoid was present in C. thomasiand C. peruviensis; the low articular con-dyles were found in C. equatoris; and a faintand shallow lingual notch between the facetsof the articular condyle was frequently pres-ent in C. equatoris. These observations sug-gest that the allocation of colombianus to theCentral American group nigrescens shouldawait a thorough revision of the South Amer-ican species.

Cryptotis peruviensis represents the south-ernmost distribution of the South Americanshort-eared shrews. As early as 1940, Ca-brera and Yepes suspected that the Huanca-bamba Depression and its arid habitats im-peded distribution of the genus Cryptotis fur-ther south, and until now that appeared to be

1997 I1I

AMERICAN MUSEUM NOVITATES NO. 3202

Fig. 5. Collection localities in Colombia, Ecuador, and Peru for specimens of Cryptotis examined.

12

VIVAR ET AL.: NEW SPECIES OF CRYPTOTIS

so. Several other taxa (e.g., Caenolestes spp.,Tapirus pinchaque) do not occur south of theHuancabamba Depression, indicating thatthis geographic feature might be an impor-tant barrier for north-south dispersion. How-ever, patterns of mammal distribution innorthern Peru, especially of nonvolant spe-cies, are still vaguely known. Tentative sce-narios of the role of the Huancabamba De-pression on bat distribution (Koopman, 1978;Pacheco and Patterson, 1992; Patterson et al.,1992) concur in reporting that the Huanca-bamba Depression has been of no major sig-nificance in impeding north-south dispersion.Interestingly, it was concluded that forhigh-elevation bird species, the Huancabam-ba Depression is not an important barrier fordispersal either (Parker et al., 1985). The im-portance of the Huancabamba Depression asa barrier for the distribution of Caenolesteswas suggested by Albuja and Patterson(1996). However, all investigators agree thatthe distribution patterns for shrew-opossum,shrews and other non-volant cloud forestsand paramo mammalian species await furtherstudy. It should be emphasized that suitablehabitats north and south of the HuancabambaDepression have not been extensively sam-pled and what may appear to be patternsmight turn out to be simply collecting biases.

For example, the deer Mazama rufina wasonce considered to be limited to the northernAndes, but it has been recently reported inLambayeque, south of the Huancabamba De-pression and on the western side of the An-des (Pacheco et al., 1995).The Cajamarca and Piura departments

where Cryptotis peruviensis has been discov-ered are unfortunately among the mostdensely populated, which correlates withhigh rates of habitat deterioration. Virtuallyno national parks exist in this area. The onlyprotected area is the Tabaconas-NamballeNatural Sanctuary, and the Cutervo NationalPark, both too small to be considered repre-sentative of the region. The Cajamarca de-partment is the only region in Peru wherenorthern and southern Andean elements arepresent in a single geopolitical region. TheCajamarca department is practically dissect-ed by the Rio Chamaya that flows eastwardto join the Rio Maranoin, giving this area aunique importance for systematics, biogeog-raphy, and conservation. The Cajamarca de-partment, and northern Peru on a broaderscale, should have highest priority for diver-sion of major conservation efforts. Its uniquehabitats, the extensive and increasing habitatdeterioration, and the meager amounts ofcurrent research and conservation effortsshould be addressed.

REFERENCES

Albuja, L., and B. D. Patterson1996. A new species of northern shrew-opos-

sum (Paucituberculata: Caenolestidae)from the Cordillera del C6ndor, Ecua-dor. J. Mammal. 77: 41-53.

Anthony, H. E.1921. Preliminary report on Ecuadorian

Mammals, No. 1. Am. Mus. Novitates20: 5 pp.

Brack-Egg, A.1986. Ecologia de un pais complejo. In Gran

geograffa del Peru 2: 175-319. Madrid:Ed. Manfer-Mejia Baca.

Cabrera, A.1958. Catailogo de los mamfferas de America

del Sur. I (Metatheria-Unguiculata-Car-nivora). Rev. Mus. Argentino Cienc.Nat. "Bernardino Rivadavia" 4: 307pp-

Cabrera A., and J. Yepes1940. Mamiferos Sud-Americanos (vida, cos-

tumbres y descripcion). Historia Natu-ral Ediar. Buenos Aires: Companifa Ar-gentina de Editores.

Choate, J. R.1970. Systematics and zoogeography of Mid-

dle American shrews of the genusCryptotis. Univ. Kansas Publ. Mus.Nat. Hist. 19: 195-317.

Choate, J. R., and E. D. Fleharty1974. Cryptotis goodwini. Mamm. Species

44: 3 pp.Corbet, G. B., and J. E. Hill

1991. A word list of mammalian species. Ox-ford: Oxford Univ. Press.

Goldman, E. A.1912. New mammals from eastern Panama.

Smithson. Misc. Collect. 60: 1-17.Honacki, J. H., K. E. Kinman, and J. W. Koeppl

1982. Mammal species of the world. Law-rence, KS: Allen Press and The Asso-ciation of Systematics Collections.

1997 I13

AMERICAN MUSEUM NOVITATES

Hutterer, R.1986. Sudamerikanische Spitzmause: Cryp-

totis meridensis und C. thomasi als ver-

schiedene Arten. Z. Saugetierk. 51,Sonderheft: 33-34.

1993. Order Insectivora. In D. E. Wilson andD. M. Reeder (eds.), Mammal speciesof the World, 2nd ed., 69-130. Wash-ington D. C.: Smithson. Instit. Press.

Koopman, K.1978. Zoogeography of Peruvian bats with

special emphasis on the role of the An-des. Am. Mus. Novitates 2651: 33 pp.

Merriam, C. H.1897. Description of five new shrews from

Mexico, Guatemala and Colombia.Proc. Biol. Soc. Washington 11: 227-230.

Nowak, R. M.1991. Walker's mammals of the world. Fifth

Ed, vol.1. Baltimore: Johns HopkinsUniv. Press.

Pacheco, V., and B. D. Patterson1992. Systematics and biogeographic analy-

ses of four species of Sturnira (Chirop-tera: Phyllostomidae), with emphasison Peruvian forms. Mem. Mus. Hist.Nat., UNMSM (Lima) 21: 57-81.

Pacheco, V., H. de Macedo, E. Vivar, C. Ascorra,R. Arana-Cardo, and S. Solari

1995. Lista anotada de los mamiferos peru-

anos. Occas. Pap. in Conserv. Biol. 2:35 pp.

Parker, T. A. III, T Schulenberg, G. Graves, andM. Braun

1985. The avifauna of the Huancabamba re-

gion, northern Peru. In Neotropical Or-nithology. Ornithol. Monogr. 36: 169-197.

Patterson, B. D., V. Pacheco, and M. V. Ashley1992. On the origins of the Western slope re-

gion of endemism: systematics of fig-eating bats, genus Artibeus. Mem. Mus.Hist. Nat., UNMSM (Lima) 21: 189-205.

Pritchard, P. C. H.1994. Comment on gender and declension of

generic names. J. Mammal. 75: 549-550.

Ride, W. D. L., C. W. Sabrosky, G. Bernardi, andR. V. Melville (eds.)

1985. International code of zoological no-

menclature, 3rd. ed. Berkeley: Univ.California Press.

Smithe, F B.1975. Naturalist's color guide. New York:

Am. Mus. Nat. Hist.Steppan, S. J.

1995. Revision of the Tribe Phyllotini (Ro-dentia: Sigmodontinae), with a phylo-genetic hypothesis for the Sigmodonti-nae. Fieldiana Zool., n. ser. 80: 112 pp.

Stone, W.1914. On a collection of mammals from Ec-

uador. Proc. Acad. Nat. Sci. Philadel-phia 66: 9-19.

Tate, G. H. H.1932. Distribution of the South American

shrews. J. Mammal. 13: 223-228.Thomas, 0.

1912. Three small mammals from SouthAmerica. Ann. Mag. Nat. Hist., ser. 8,9: 408-410.

1921. New Cryptotis, Thomasomys and Ory-zomys from Colombia. Ann. Mag. Nat.Hist., ser. 9, 8: 354-357.

Tosi, J. A., Jr.1960. Zonas de Vida Natural en el Peru'. Inst.

Interam. Cienc. Agric. OEA Zona An-dina. Bol. Tec. 5: 271 pp.

Woodman, N.1993. The correct gender of mammalian ge-

neric names ending in otis. J. Mammal.74: 544-546.

1995. Morphological variation between Pleis-tocene and recent samples of Cryptotis(Insectivora: Soricidae) from the Yu-catan Peninsula, Mexico. J. Mammal.76: 223-231.

1996. Taxonomic status of the enigmaticCryptotis avia (Mammalia: Insectivora:Soricidae), with comments on the dis-tribution of the Colombian small-earedshrew, Cryptotis colombiana. Proc.Biol. Soc. Washington 109: 409-418.

Woodman, N., and R. M. Timm1992. A new species of small-eared shrew,

genus Cryptotis (Insectivora: Sorici-dae), from Honduras. Proc. Biol. Soc.Washington 105: 1-12.

1993. Intraspecific and interspecific variationin the Cryptotis nigrescens speciescomplex of small-eared shrews (Insec-tivora: Soricidae), with the descriptionof a new species from Colombia. Field-iana Zool., n. ser. 74: 30 pp.

14 NO. 3202

VIVAR ET AL.: NEW SPECIES OF CRYPTOTIS

APPENDIX

List of collection localities and specimens stud-ied in this report. Specimens marked with an as-terisk (*) were measured. Those marked with (#)are holotypes examined.

Cryptotis equatoris equatorisEcuador

1. BOLIVAR, Guaranda, Sinche (AMNH:66837, *66838, *66839, *66840, *66841,*66842, *66843, 66844, *66845).

Cryptotis equatoris osgoodiEcuador2. PICHINCHA, Chinchin Cocha, 4000 m

(FMNH: 53316, 53317; not located).3. PICHINCHA, Monte Pichincha (AMNH:

*62379, *62380).4. PICHINCHA, Crater Pichincha, 4300 m

(FMNH: 53318).5. PICHINCHA, San Ignacio (AMNH: *64576,

*64582, *64583, *66249).6. PICHINCHA, Verdecocha (AMNH: *64585).

Cryptotis medelliniusColombia7. ANTIOQUIA, Valdivia, Las Ventanas, 2000m (FMNH: 69812, *69813).

8. ANTIOQUIA, Antioquia, San Pedro, 2560 m(AMNH: * 149151).

9. ANTIOQUIA, Sons6n, Paramo, 7 km E,3000-3100 m (FMNH: *69815, *69817,69818).

10. ANTIOQUIA, SE de Medellin, Las Palmas,2650 m (FMNH: *69814).

Cryptotis montivagusEcuador11. AZUAY, Besti6n, 10000 ft (3048 m)

(AMNH: *47197, *47199, *#47200,*47201).

12. PICHINCHA, Guamani, Cerro Guamanf(AMNH: 63844).

13. CHIMBORAZO, Urbina (AMNH: 64623).14. CHIMBORAZO, Lake Quinvascocha GL Ca-

jas, Cuenca Azuay, 3850 m (MECN: 138, notlocated).

Cryptotis peruviensisPeru15. CAJAMARCA, Las Ashitas, 42 km W Jaen,

3150 m (MUSM: *#8373).16. PIURA, Machete, Zapalache-Carmen trail,

2050 m (LSUMZ: *26887).

Cryptotis thomasiColombia17. CUNDINAMARCA, Bogota, San Cristobal,

2800 m (FMNH: *71031, *71032, 71033,*71034, *71036, *71037).

18. CUNDINAMARCA, Bogota, San Francisco,3000 m (AMNH: *71355, FMNH: 71023,71024, *71025, 71026, 71028, 71029).

19. CUNDINAMARCA, Bogoti, P6ramo deChoachi (AMNH: *38405).

20. CUNDINAMARCA, Bogovi, Pdramo de Bo-gota (AMNH: *37381).

1997 15

Recent issues of the Novitates may be purchased from the Museum. Lists of back issues of theNovitates and Bulletin published during the last five years are available at World Wide Web sitehttp://nimidi.amnh.org. Or address mail orders to: American Museum of Natural History Library,Department D, Central Park West at 79th St., New York, N.Y. 10024. TEL: (212) 769-5545. FAX:(212) 769-5009. E-MAIL: [email protected]

(D This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper).